Mrunmai Phatak
| Telefon: | +49 (821) 598 2462 |
| E-Mail: | mrunmai.phatak@uni-auni-a.de () |
| Raum: | 1011 (N) |
| Adresse: | Universitätsstraße 6a, 86159 Augsburg |
Publikationen
|
2026 |
|
Julian Lorenz, Vladyslav Kovganko, Elias Kohout, Mrunmai Phatak, Daniel Kienzle and Rainer Lienhart. in press. DSFlash: comprehensive panoptic scene graph generation in realtime. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026), June 3-7, 2026, Denver, CO, USA. IEEE, Piscataway, NJ, 17388-17398 DOI: 10.48550/arXiv.2603.10538 |
|
Roman Küble, Marco Hüller, Mrunmai Phatak, Rainer Lienhart and Jörg Hähner. in press. Modernising reinforcement learning–based navigation for embodied semantic scene graph generation. preprint. DOI: 10.48550/arXiv.2603.25415 |
|
2025 |
|
Julian Lorenz, Mrunmai Phatak, Robin Schön, Katja Ludwig, Nico Hörmann, Annemarie Friedrich and Rainer Lienhart. in press. COPA-SG: dense scene graphs with parametric and proto-relations. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, October 19–23th, 2025, Honolulu, Hawai'i. IEEE, Piscataway, NJ, 7604-7613 |
|
Julian Lorenz, Mrunmai Phatak, Robin Schön, Katja Ludwig, Nico Hörmann, Annemarie Friedrich and Rainer Lienhart. 2025. COPA-SG: dense scene graphs with parametric and proto-relations. In Hilde Kuehne, Gerard Medioni, Dimitris Samaras, Jingyi Yu, Ramin Zabih (Eds.). 2025 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), Honolulu, HI, USA, 2025, October 19-20, 2025. IEEE, Piscataway, NJ, 7663-7672 DOI: 10.1109/ICCVW69036.2025.00788 |
|
Mrunmai Phatak, Julian Lorenz, Nico Hörmann, Jörg Hähner and Rainer Lienhart. in press. HOIverse: a synthetic scene graph dataset with human object interactions. In 2025 IEEE International Conference on Multimedia Information Processing and Retrieval (MIPR). IEEE, Piscataway, NJ |
LEHRE
- SS 2026: Multimedia Projekt
- WS 2025/2026: Multimedia Projekt
BETREUTE ABSCHLUSSARBEITEN
- Christian Spreng. Learning to imitate biometrically plausible human motions. Bachelorarbeit, Januar 2026